New immunotoxins (ITs) are being developed for clinical use. Laboratory efforts are focused on using phage display to identify new antibodies and to produce new ITs directed at various tumors (breast, ovary, prostate, lung, colon, mesothelioma, astrocytoma) and lymphomas and leukemias. 1. A mutant form of the EGF receptor is found in many glioblastomas, breast, lung and ovarian cancers. D. Bigner (Duke University) has made MAbs that recognize only the mutant receptor. Chemical conjugates of these with PE are cytotoxic to cells with mutant receptors but have low activity. Recombinant ITs with high activity could not be produced from these MAbs. Antibody phage display was used to generate a new antibody that produces a very stable and active recombinant IT which is being evaluated for clinical use. 2. Prostate cancer: MAb E4 reacts with normal prostate cells and differentiated prostate cancers. It has reactivity with other normal tissues. It has been chemically coupled to mutant PE (PE35) to make an IT that is very cytotoxic to prostate carcinoma cell lines. The genes of this antibody are being cloned to make a recombinant IT. 3. Melanoma: Melanomas and melanocytes contain a form of the MSH receptor (MCS-1) that is not present in other tissues. We are preparing antibodies to MCS-1 to examine their possible usefulness as ITs 4. Ovarian cancer: Most ovarian cancers and mesotheliomas contain a protein termed mesothelin. We are developing high affinity antibodies to mesothelin which can be used to direct toxins (or other cytotoxic agents) to these cells. We have cloned the mesothelin cDNA and created tumors that express large amounts of mesothelin. These tumors are used to evaluate targeted therapies with antibodies directed at mesothelin. Using phage display, new antibodies to mesothelin have been identified. 5. Epithelial tumors: LMB-1 has produced several clinical responses but it damages capillaries because its large size retards its exit from the circulation. LMB-7 is much smaller and damages epithelial cells of the stomach, because it rapidly leaves the circulation. We are constructing several ITs of different sizes with the goal of finding a size which will allow the IT to escape from the circulation into tumors, but will restrict access to stomach epithelial cells. We have designed ITs that do not require proteolytic processing by inserting a dsFv fragment either between domains II and III or at the end of domain III prior to the REDLK sequence, and two very active ITs were made with Fv fragments of MAbs B1 and e23 (anti-erbB2).